LangChain - ChatBot

Description

The purpose of this project is to create a chatbot that can be used to communicate with the user in a natural language

Requirements

• Ollama 0.1.25 or higher
• Python 3.12.2 or higher
• pip 23.2.1 or higher

(Optional) Python Virtual Environment

It is recommended to use a virtual environment to manage the dependencies of the project.

This will allow you to install the dependencies without affecting the global python installation.

To create a virtual environment, run the following command:

python -m pip install virtualenv
python -m venv .env

Ollama Installation

This repository uses the Ollama library to create the chatbot.

1. Install based on your OS https://ollama.com/
2. Install a language model and automatically run it

   #syntax: ollama run <model name>
   
   ollama run llama2
   

(Optional) WSL changes needed for external access

If you are using WSL, you will need to make some changes to the Ollama installation to make accessible outside the WSL environment.

1. Open the Ollama service file

   sudo vim /etc/systemd/system/ollama.service
   

2. Locate the [Service] section and then Environment.

   At the end of the line set the OLLAMA_HOST variable to listen on all interfaces.

   Environment=<...> "OLLAMA_HOST=0.0.0.0"
   

3. Reload the system services

   sudo systemctl daemon-reload
   

4. Restart the Ollama service

    sudo systemctl restart ollama
   

5. Get your WSL ip address

   ip addr show eth0
   

6. Open powershell in administrator mode and run the following command to open the port

   netsh interface portproxy add v4tov4 listenport=11434 connectport=11434 connectaddress=<WSL_IP>
   

Verify Ollama Service is Running

Open a web browser and navigate one of the following URLs:

• http://localhost:11434

• http://<WSL_IP>:11434

---

Project Setup

1. Clone the repository.

2. Install dependencies using requirements.txt

   pip install -r requirements.txt
   

3. In the root of the project, create a file called .env and add the following environment variables:

   Note: replace with the IP address of the Ollama service

   OLLAMA_URL=http://<IP>:11434
   

4. Run the test file to verify your code can communicate with the Ollama service

   python hello_ollama.py
   

5. Run the main file to start the chatbot

   python chatapp.py
   

6. Exit the chatbot by typing "exit" into the user input field